We’ve mentioned a few times over the last couple years our ambition to roll out bracketology models (and similar concepts) for more college sports, so with the women’s soccer season starting today, we wanted to give those of you who’ve been following that effort an update.
The models aren’t here.
But they’re (hopefully) on their way.
One problem with college sports is that once you get to a certain point in the financial pecking order, nobody’s keeping databases of past scores. They’re available, but you have to cobble together the database yourself, and we don’t have much experience with data scraping. We are, with the help of a reader and friend down in Austin, working on the ones for soccer and women’s volleyball, but we don’t have any of those database assemblers fully functional yet, so we’re a decent ways away from having a bracketology model for any of those three sports.
Meanwhile, football’s bearing down on us. Week Zero is nine days away. We’re trying to bring back our FCS model this year, pairing it with our College Football Playoff probabilities, and we’re trying to add bowl projections and more accurate conference tiebreakers while increasing interactivity and doing enough back-testing to convince skeptics of the model’s accuracy, which is a lot. We might not get it all there, and adding the FCS piece has required us to do a lot more database verification than we’d have to do for only an FBS model (long explanation on that). Thankfully, our CFP probabilities are good already, so the bones are there, but we’ve got a lot of work to do between now and next Friday morning, our current target for launching the damn thing.
So, for the time being, football’s the focus. When we’ve got that up and moving, we’ll return our attention to soccer and volleyball, though the approach of the NHL season and college basketball may push those to the shelf again until next summer. We’ll keep you posted.
hahahaha
push.